this worked! Thank you so much!
1 Like
In my case，I tried restarting the computer, logging into my Epic account again, and switching networks. I eventually succeeded, but I’m not sure which step actually worked. I kept trying to reinstall the plugin, and it suddenly worked in the end. Nothing was modified, only restarted and switched networks.
Does not work. New Installation (standard didn’t mess with anything) with latest 5.8 installed and options Meta Humman Creator Core Data applied. But I get this message:
Tried suggested methods for solution from this thread but with no success. Also there is an update button in the popup after clicking on installed plugins, but it doesn’t do anything.
Nvidia Driver Version: 610.62
1 Like
This absolutely corrected the issue for me. Thanks!
I would love the ability to create mocap from multiple camera angles. The single camera is just not good enough. When I tested it, it looks fine from one angle, but as you rotate around, there are things that are off because the single camera cannot capture the depth information.
I reset the settings in both the launcher and downloader, and it worked for me because the file wasn’t in the Epic Launcher folder.
I use Linux prebuilt binary and this plugin compiles correctly. However, I keep getting runtime error and after doing AI analysis, it said that the plugin is not supported for Linux, so is this plugin not supported yet for Linux? I use the Fab plugin to download and install and it did say this may not be compatible with current OS
UE 5.8 MetaHuman Markerless Motion Capture - “Capture data frame rate is zero”
I’m using Unreal Engine 5.8 with the MetaHuman Animator Markerless Motion Capture plugin.
My workflow:
Imported an MP4 (H.264, 30 FPS, constant frame rate).
Mono Video Ingest completes successfully.
Capture Data asset is created.
JPG image sequence is generated correctly (1920×1080).
However, when I assign the Capture Data to a MetaHuman Performance asset, I get:
Invalid Frame Rate Error
Selected capture data frame rate or image/depth sequence has invalid frame rate.
The Output Log shows:
Capture data frame rate is zero.
Image sequence frame rate numerator equal zero.
SetFootageCaptureData: footage capture data has invalid frame rate; assignment rejected.
Also, the generated Img Media Source shows:
Player: None
Resolution: 2×2
Frame Rate: 0
Even though the extracted JPG sequence exists and is valid (1920×1080).
Has anyone encountered this? Is this a known UE 5.8 Markerless Motion Capture bug, or is there a fix?
For anyone concerned about the jittering : There’s this variable in the plugin
static constexpr int InnerWindow = 192;
that controls processing chunks and at their borders there’s always pose snapping. Increasing this to the number of frames of my animation (like 1000) makes the snapping go away. I also added rolling average smoothing as an animation modifier after the animation export because only the facial expressions get smoothing by default. I wish these settings would be exposed/builtin so I don’t have to keep custom modifications around.
I also noticed that in large animations like 20 000 frames it gets to ~340GB committed memory and then dies due to OOM ( I only have 96GB), but at least I can process chunks safely of ~3000 frames by modifying start/end frames but doing that automatically inserts one of those windows that introduce snaps at the start and end of the chunk.
2 Likes
Very nice tool! I understand is a new product and is prone to error but I wasted 3 hours on a 30 second video process just for the final result to be a moving head, even tho I followed absolutely every step, got full body video with good lighting setup.
The template default skeleton at the beginning is not showing, and I firmly believe this is the root cause
[UE 5.8.2] MetaHuman Markerless Motion Capture Stage 4 causes GPU reset and black screens on RTX 5070
Hello,
I am experiencing a repeatable GPU/display crash when using the experimental MetaHuman Animator Markerless Motion Capture plugin in Unreal Engine 5.8.2.
System specifications:
- Windows 11
- Unreal Engine 5.8.2, build 56702186
- MetaHuman Animator Markerless Motion Capture plugin 1.0.0 Beta
- NVIDIA RTX 5070, 12 GB VRAM
- AMD Ryzen 9 9950X3D
- 32 GB system RAM
- NVIDIA Studio Driver 616.56
- DirectX 12
Source video:
- H.264 MP4
- 1080 × 1920
- 30 FPS
- Approximately 7 seconds
Problem:
Processing completes Stages 1–3 normally. As soon as Stage 4 begins, all connected monitors lose signal and become completely black. The computer appears to remain powered on and working, but the display never recovers. I eventually have to force the computer to shut down.
This has happened repeatedly, including with a very short video.
The Unreal log reports ONNX Runtime/DirectML errors during Stage 4, followed by a GPU device reset/removal:
DXGI_ERROR_DEVICE_RESET
DXGI_ERROR_DEVICE_REMOVED
HRESULT: 887A0005
onnxruntime.dll
DmlExecutionProvider
ExecutionProvider.cpp(952)
This does not appear to be a normal VRAM out-of-memory error. At the time of the failure, approximately 6.9 GB of VRAM was being used, with an approximately 11.2 GB budget. The Unreal crash information also reported bIsOOM=0.
Troubleshooting already performed:
- Clean installation of NVIDIA Studio Driver 616.56
- The problem also occurred with NVIDIA driver 595.76
- Restarted Windows after driver installation
- Removed custom TdrDelay and TdrDdiDelay registry values
- Disabled ray tracing
- Turned off the 4K monitor and tested using only a Full HD monitor
- Tested body tracking with face tracking disabled
- Tested only 30 frames
- Disabled the preview pass
- Used a fresh temporary performance asset
- Confirmed DirectX 12 is enabled
- Confirmed the problem still happens with the short 7-second clip
The 30-frame body-only test produced the same DirectML error at Stage 4 and reset the display driver.
Comparison:
The markerless body-capture workflow completed successfully on another computer using:
- Unreal Engine 5.8.0
- RTX 5090
- Ryzen processor
- 128 GB RAM
Therefore, I suspect a compatibility issue involving one of the following:
- MetaHuman Animator Markerless Motion Capture plugin and UE 5.8.2
- ONNX Runtime/DirectML and the RTX 5070
- A regression between UE 5.8.0 and UE 5.8.2
- Different memory requirements or plugin builds
Has anyone reproduced this on an RTX 5070 with UE 5.8.2?
Is there a known issue or workaround for the Stage 4 DirectML error 887A0005? Is it possible to test the UE 5.8.0 version of the plugin without compiling the complete engine from source?
I can provide the complete Unreal log if required.
Hooray, Windows 11 update apparently turned on Smart App Control automatically and it now blocks UnrealEditor-Segmentation.dll.
The plugin binaries are not signed, so it won’t allow the plugin to run without turning off SAC completely (which apparently can only be turned on again by reinstalling Windows?).
Please get the binaries signed, so this won’t be an issue going forward.
Have you found a fix? I have been suffering with this for a month now. Can you please reply!
Have you solved the problem? Been trying to look for a solution for months now, nothing works. But somehow works in version 5.6 (I don’t know how or what I did) but doesn’t work on version 5.8